Famed AI Researcher Launches Controversial Startup to Replace All Human Workers Everywhere (techcrunch.com) 166
TechCrunch looks at Mechanize, an ambitious new startup "whose founder — and the non-profit AI research organization he founded called Epoch — is being skewered on X..."
Mechanize was launched on Thursday via a post on X by its founder, famed AI researcher Tamay Besiroglu. The startup's goal, Besiroglu wrote, is "the full automation of all work" and "the full automation of the economy."
Does that mean Mechanize is working to replace every human worker with an AI agent bot? Essentially, yes. The startup wants to provide the data, evaluations, and digital environments to make worker automation of any job possible. Besiroglu even calculated Mechanize's total addressable market by aggregating all the wages humans are currently paid. "The market potential here is absurdly large: workers in the US are paid around $18 trillion per year in aggregate. For the entire world, the number is over three times greater, around $60 trillion per year," he wrote.
Besiroglu did, however, clarify to TechCrunch that "our immediate focus is indeed on white-collar work" rather than manual labor jobs that would require robotics...
Besiroglu argues to the naysayers that having agents do all the work will actually enrich humans, not impoverish them, through "explosive economic growth." He points to a paper he published on the topic. "Completely automating labor could generate vast abundance, much higher standards of living, and new goods and services that we can't even imagine today," he told TechCrunch.
TechCrunch wonders how jobless humans will produce goods — and whether wealth will simply concentrate around whoever owns the agents.
But they do concede that Besiroglu may be right that "If each human worker has a personal crew of agents which helps them produce more work, economic abundance could follow..."
Does that mean Mechanize is working to replace every human worker with an AI agent bot? Essentially, yes. The startup wants to provide the data, evaluations, and digital environments to make worker automation of any job possible. Besiroglu even calculated Mechanize's total addressable market by aggregating all the wages humans are currently paid. "The market potential here is absurdly large: workers in the US are paid around $18 trillion per year in aggregate. For the entire world, the number is over three times greater, around $60 trillion per year," he wrote.
Besiroglu did, however, clarify to TechCrunch that "our immediate focus is indeed on white-collar work" rather than manual labor jobs that would require robotics...
Besiroglu argues to the naysayers that having agents do all the work will actually enrich humans, not impoverish them, through "explosive economic growth." He points to a paper he published on the topic. "Completely automating labor could generate vast abundance, much higher standards of living, and new goods and services that we can't even imagine today," he told TechCrunch.
TechCrunch wonders how jobless humans will produce goods — and whether wealth will simply concentrate around whoever owns the agents.
But they do concede that Besiroglu may be right that "If each human worker has a personal crew of agents which helps them produce more work, economic abundance could follow..."
Cool (Score:3)
>enable the full automation of all work
>we're hiring
I wonder for how long.
Re:Cool (Score:5, Insightful)
My guess until they have embezzled enough "investor" money or got bought by somebody. Call it 3-5 years. Maybe. The claims made are insane and come late in the game, so this scam may simply collapse quietly.
Then what? (Score:5, Insightful)
Re: Then what? (Score:5, Insightful)
How about basic income?
Re: Then what? (Score:3)
Re: Then what? (Score:3)
Why not print money faster than prices rise, and think of prices in terms of purchasing power units which would remain stable even if hyperinflation raged on?
If you start from Fischer Black's observation that prices are noise, why does inflation matter when we can adapt to it using indexation?
Re: (Score:2)
How about just outlaw price rises. That would be just as easy.
Re: (Score:2)
Yep, that's exactly my point.
The idea of raising "basic income" faster than inflation, is like a Ponzi scheme paying out higher and higher returns, when people start to figure out the money isn't there. Basic income *is* at its core, a Ponzi scheme.
Re: (Score:2)
If it's all laid out openly at the start, where's the fraud? How long has the market "Ponzi scheme" been enriching investors by increasing stock returns faster than inflation, creating a defacto private version of indexation?
Re: (Score:2)
For a satirical take on hyperinflation, I recommend Douglas Adams' novel "Restaurant at the End of the Universe", in which near the end, the occupants of a crashed spaceship try to function in an undeveloped "natural" environment. They end up adopting leaves as legal tender. They stuffed their suits with them until they bulged. Then they brilliantly figured out how to solve the fact t
Re: Then what? (Score:3)
Re: (Score:3)
If we were fully automated, you would have to come up with a different way to go about distributing resources, as we wouldn't be earning an income anymore. I could see aspects of that working out just fine but the big elephant in the room is how do we choose who gets to live where if we all get equal access to resources given everything is automated?
Since raw materials are always finite, things would still have a "cost". This is where individuals get to choose how to spend their allotted resources.
It would
Re: (Score:2)
No way to get ahead? FUCK THAT!
I've been in this employment/rewards game since the 70's, there is no getting ahead. Inflation, recession, stagflation, its all economic gambling, spin the wheel, take a chance, too bad kid you lost your life/savings better luck next time. I'm retired and watching it all evaporate and hoping I don't have to eat cat food. The cake is a lie!
Maybe the blockchain, or AI, help me Obi Wonna Tarrifinski.
Re: Then what? (Score:2)
We would have to elimate most private property from our society in order to redistribute the productivity of robots and AI agents to the human beings that need to live, eat, breath, etc.
Allowing capitalism and the price system to reach it's logical conclusion with zero human labor is not going to function. There would be a violent uprising when enough people had no job and no money. And the property owners would spend most of their gains hiring private security.
Re: Then what? (Score:2)
I had wierdly techno-utopian idea a long time ago
Re: Then what? (Score:5, Insightful)
The problem is: how do we get there from here? Powerful corporations and wealthy individuals have an incredible amount of influence, and they are not going to agree to a heavy tax on their profits. Besides, countries that are the first to implement this might see their companies fleeing to greener pastures.
UBI doesn't work (Score:2)
I don't think we're actually going to get our way out of this, I seriously think the human race is doomed. Watching Donald Trump win and project 2025 being implemented combined with the left wing and the centrists basically bickering among themselves while fapping about had stupid protests has pretty much blown any hope I had left.
Re: (Score:3)
The premise that AI and bots can replace all human work, is nothing more than a marketer's daydream, or a storyteller's nightmare. We've been automating work for thousands of years. We've automated more than 90% of all farm work, all blacksmith work, all carriage driver work, all scribe work, and yet, pretty much everybody who wants a job, at least in the US, has one. We're not going to finish automating all work, ever.
On the other hand, the premise that basic income can exist, or that it can solve the worl
Re: (Score:2)
How about basic income?
We will need an UBI you can live off decently (!) because wealth distribution is slowly becomming too broken to be sustainable. Even the rich want a society that somewhat functions. But most people (about 80% in most UBI studies) want and need work. Hence there is a whole secondary problem to this.
Re: Then what? (Score:5, Informative)
You say that like a bad thing. That isn't failure, it's liberation achieved via automation. As technology grows increasingly capable [mckinsey.com] of handling routine tasks across sectors, freeing human potential this freedom allows focus on creativity, learning, community, and well-being. Evidence from UBI pilots [stanford.edu] shows people often invest in education, health, and small ventures, not just passive leisure. This aligns with known drivers of intrinsic motivation [selfdeterm...theory.org] beyond pure economic necessity. The argument about mass idleness or loss of purpose is weak; purpose isn't and shouldn't be solely derived from paid work [philosophersmag.com], and pilot data show minimal labor supply reduction [helsinki.fi], refuting predictions of widespread apathy. Human value exceeds narrow economic output metrics [oecdbetterlifeindex.org] like GDP [bu.edu]. [warning: PDF] While there are many people who argue that failing to work is unethical, I've seen no argument for this point of view other than an appeal to consequences. Let me counter with the famous essay, In Praise of Idleness [harpers.org]
Mechanisms like UBI are vital to channel automation's productivity gains, ensuring shared prosperity [rooseveltinstitute.org] [Warning: PDF] over mass poverty, and there is nothing wrong with the resulting idleness. Not in act, nor consequence.
As for evidence the idleness won't happen like you expect, study after study [basicincome.org] has shown it will not. That recent one has weaknesses, such as being only a 3-year pilot, but it provides significant evidence contradicting common criticisms of Universal Basic Income. The research, conducted by DIW Berlin and Mein Grundeinkommen, found that recipients of EUR 1,200 monthly did not withdraw from the labor market or significantly reduce working hours compared to a control group, directly challenging the "social hammock" stereotype. Beyond the lack of negative work effects, the study also documented substantial improvements in recipients' mental health (comparable to therapy), life satisfaction, and sense of purpose, alongside responsible financial behaviors like saving and increased altruism.
This transition represents progress: leveraging technology to enable widespread human flourishing and choice, leisure included, beyond the constraints of forced labor. Mods, you may now moderate, as always, by whether you like UBI rather than the quality of this post.
Re: Then what? (Score:2)
Re: (Score:3)
It goes even deeper than poverty or richness, it's control, complete, absolute control.
If you're the sole provider of all goods, you can decide what can even be sold or not.
An corporate emperor of mankind that does not like strawberry will force it to completely disappear from existence, and force his taste on the entire world.
And if you try to make it yourself? you and your whole family gets kicked out of the UBI, because no one will accept to pay it without having advantages.
Re: (Score:3)
Your mistake in reasoning here is in understanding the nature of power. Money is not the same as power. Neither is position. Both money and position depend on a society willing to afford power to those things. As we seen, all it takes is a small minority rejecting either or both to break the system, rendering the formerly powerful powerless -- and sometimes headless.
True power comes from the collective action of the masses. Money only has value because we all agree that money has value. Our elected of
Re: (Score:2)
Money is a proxy for power - in a lawful society.
In a dictatorship or lawless society, power comes from the barrel of a gun or the head of a sledgehammer.
Ask Jack Ma.
Re: (Score:2)
With full automation, you don't need or have any kind of copyright or legal DRM of any kind. Those exist so people can generate income off their works. Without the need for income, you don't need to generate it. Anyone could request the automation make them a thing that bob got and they could. If anything, the way to gain status in such a society would be to actually create something that went viral or otherwise let people do something new.
You could of course still create whatever you want from scratch or w
Re: (Score:2)
You have never had a hobby?
Re: (Score:2)
Sure. My hobbies don't produce anything. They are just fucking off. Scuba diving, shooting pool, etc.
Re: (Score:2)
I garden a bit amongst other stuff, which does happen to produce something though that's not why I do it. Still given more free time, I'd spend more time doing it. Perhaps with more free time you can do the stuff you do and even find a hobby that does produce something if that is important to you.
Re:Then what? (Score:5, Insightful)
In principle it could be a very beneficial revision of the current paradigm, where the majority of the population works an unsatisfactory job until they retire or die. You could actually do things you want to do with your life, imagine that.
"If each human worker has a personal crew of agents which helps them produce more work, economic abundance could follow..."
I'm skeptical but I like the concept. The value of labor would plummet since it would mostly be done relatively cheaply with machines. And if the machines are cheap, everyone could just have them produce whatever we need or desire. Science fiction is full of imaginings like this.
Re: Then what? (Score:2)
Re: Then what? (Score:5, Interesting)
>> a hard row to hoe though
True, but it could eventually be difficult to control the automated production. Its already possible to roll your own AI these days, but in many cases it isn't even worth the trouble. I've got access to multiple AI coding assistant backends right now for a monthly subscription fee of $10. They are literally competing for my attention, I'm getting offers for freebies. Meanwhile my productivity has zoomed. Now my time is mainly spent figuring out exactly what I want it to do, and did it accomplish that to my satisfaction.
According to wikipedia;
The labor theory of value (LTV) is a theory of value that argues that the exchange value of a good or service is determined by the total amount of "socially necessary labor" required to produce it.
When that socially necessary labor falls off to near zero the value of the good or service should likewise decline. I don't know who gets rich in a scenario like that.
Re: (Score:2)
You'll still need a currency unless you find a way to convert energy it into anything you want. If you can "magic" raw resources into existence, then sure, goods and services could fall to zero. Since resources are finite, we'll still need a currency because things will still need a "cost" in terms of actual raw materials used.
In my mind, if everything is automated, we all would need to be given the exact same spending power, regardless of anything you spend your time doing. You ban any other source of inco
Re: (Score:2)
We would still be resource-constrained I agree. But it seems like the time and difficulty of creating things from basic resources could be vastly reduced at some point over the next few years. Maybe that will make the energy requirements easier to do as well.
I'm trying to be optimistic. A good situation could happen for us.
Re: (Score:2)
I think that's wrong - value derives from demand for a product, irrespective of the labor required to produce it. I could spend decades crafting an exquisite dryer lint sculpture. If no one wants it, the amount of labor spent on it doesn't matter.
Conversely, I could spend an hour a day decry
Re: (Score:2)
That is all true, but nobody is going to build a long-term business that sells goods for less than it costs to produce them. (Unless you subsidize it with advertising, but in that case your product is the consumer and you're selling to the advertiser)
Re: (Score:2)
Those controlling the automated production would have to be nice enough to give away the fruits of that production for free.
What makes you think they'll have a choice?
Re: Then what? (Score:2)
Re: (Score:2)
Massive robot wars will replace footbal as entertainment. "Tonight, we have Russian vs Chinese robots fighting over Eastern Siberia .. "
Re: (Score:2)
Once every single job has been replaced and every single human is completely out of work, what do we do? How is society supposed to continue to function? Are we going to have some kind of massive revolution where we abolish currency or something?
This is indeed the question we need to answer.
This company that makes that declaration there will be no humans working right out loud and in public makes the question unavoidable, but it applies to all of the AI/robotics companies. How does our economy, which depends on humans working, function with no humans working?
Re: (Score:2)
Let's not get ahead of ourselves, you are buying into the hype. You do realize this is a make-believe sales pitch, right? No product ever, has actually lived up to the hype produced by its marketing team.
Re: (Score:2)
Shsssh. It's like pro-wrestling. We know it's fake but we want to believe! Just let us have this moment where we dream about utopia.
Re: (Score:2)
But but but...pro-wrestling is all real!!!
Good point.
Techno feudalism (Score:4, Interesting)
And then 99.5% of the human population lives in a staggering level of abject poverty. The kind that say native Americans stuffed on the reservations live under. If you get uppity and demand better they use drones to kill you. You're allowed to have small arms and weapons and fight amongst yourself but you'll never be given access to enough weaponry to fight the drones and the actual military that's keeping you down.
Modern weapons Technology means revolutions can't happen anymore so we're trapped in a horrifying dark age until eventually one of the idiots running the world fucks up and launches the old nukes and kills off the human race.
Except for the part where we all die from nukes at the end that's basically what guys like Peter thiel and Elon Musk have in store for us.
Re: (Score:2)
Don't worry. It cannot be done at this time. Maybe in 100 or 1000 years. Maybe never. Scams like this (and it is a scam) run generally on most people not even beginning to understand the complexities involved in making something like this actually possible.
Too broad, not specific enough. (Score:3)
I tried to find out what they are going to do exactly, and this is the only relevant part of their statement:
"Today we’re announcing Mechanize, a startup focused on developing virtual work environments, benchmarks, and training data that will enable the full automation of the economy.
We will achieve this by creating simulated environments and evaluations that capture the full scope of what people do at their jobs. This includes using a computer, completing long-horizon tasks that lack clear criteria for success, coordinating with others, and reprioritizing in the face of obstacles and interruptions."
The Rich Will Make The Poor Illegal (Score:3)
That is the plan anyway. Except AI and robots aren't going to cut it because they can't even cut the grass or wash the dishes in the sink.
What a dilemma. Stupid, dirty poor people!
Re: (Score:2)
I thought the plan of the rich was to increase their wealth on the backs of the poor. Have their tactics changed while nobody was looking?
Re: (Score:2)
Funny how everybody wants to be richer. But the "rich" are by definition, "evil." I wonder what annual income, specifically, divides the "evil rich" from the "downtrodden poor." I guess we should be careful not to try to make too much money, so we don't overshoot that threshold.
Oh but hold on, by the standards of many countries, *YOU* and *I* are rich. I guess we're already evil.
Re: (Score:2)
The reason that the "rich" are often considered evil is that often it takes being a psychopath to get "rich". What kind of person is not happy having multi-millions and needs billions? There are people who, for example inherit wealth and then give much of it away or do other things with it to benefit others instead of increasing their wealth.
Studies have shown that there's a certain wealth that gives maximum happiness to most people. It varies on location but might be an income of $100,000. Basically enough
I know exactly who'll be enriched (Score:5, Insightful)
"Besiroglu argues to the naysayers that having agents do all the work will actually enrich humans, not impoverish them, through "explosive economic growth."
I know exactly who this will enrich, and I guarantee you it won't be "all humans". If you think Bezos, Zuck, Musk are pricks now, just wait until you see the next set of bro trillionaire AI bosses. Gutter level misery for most, privileged lives for few.
Re: (Score:3)
It's already been explored by Marshall Brain's Manna [marshallbrain.com].
The complete text is there, basically going through what happens when the world is automated away and basically everyone is stuffed into tiny buildings that the owners are forced to provide so they could have their wealth.
It even shows how insidious it is - starting at the bottom. The only reason it hasn't happened yet is that automation at fast food joints hasn't reduced employment but increased it as the automated kiosks and apps have been encouraging m
Re: (Score:2)
How can you have an economy increase when the exchanges are what defines the growth?
So bot-to-bot exchanges of credits will be counted? You can't have unlimited credits for everybody. At some point most people will level off their consumption but for most the planet we lack the level of resources for that to happen.
So everybody gets a car... we don't have the resources for that to happen and imagine how much worse traffic would be. why are you traveling in the 1st place? to go to work? to avoid using deliv
Post-scarcity (Score:5, Interesting)
The natural thing to think about with something like this is the hypothetical post-scarcity society. Unfortunately, even if we could somehow get it to work out, there would still be such a big disparity that most people would just live in squalor. Then there's the inevitable brain rot that will occur, whereby people just sit around and reproduce and develop or maintain no skills at all. So when the solar flare hits we go back to prehistoric levels.
Re: (Score:2)
That sounds contradictory. If we achieve post-scarcity, how would "most people" just live in squalor? If they remain poor, then we clearly would not have achieved "post-scarcity."
Re: (Score:2)
That sounds contradictory. If we achieve post-scarcity, how would "most people" just live in squalor? If they remain poor, then we clearly would not have achieved "post-scarcity."
It is, which is my point. The means of production then changes to whomever is owning the AI. Since there is no need for the workers (i.e., "most people") they are basically thrown to the wolves. In fact, they're a net drain, so it turns into some dystopian world like in Elysium.
Re: (Score:2)
You said, "Unfortunately, even if we could somehow get it to work out..."
If there are still poor people, then we wouldn't have "somehow" been able "get it to work out." So whatever follows your "unfortunately" is moot.
In your dystopian world, all these "poor people" wouldn't just sit and rot, they would look for ways to survive. And in the process, they would find inventive solutions to survive and thrive, that are outside the realm of the "rich people's" system. People, even poor people, aren't helpless. T
Re: (Score:2)
Then there's the inevitable brain rot that will occur
That is incredibly unlikely.
Re: (Score:3)
You could always require 10 hours of community service from everyone, every week. If you choose to study something, you could contribute to that field.
I'm sure scientist would love to just be able to do research without worrying about making real money or whether they would get funding.
There is still "fun" work that could be done and I think people would generally want to do stuff. Right now, we do what we have to do to survive. It doesn't always align with what we wish we were doing. In some cases, what we
Re: (Score:2)
As studies of UBI shows, you are in a pretty small minority (10-20% of all people), the rest would continue to work even of there was no pressure to do so. They would be more selective. But there are hughe organizations out there running mostly on volunteer work.
All the same (Score:4, Informative)
Every time someone writes an article about "AI," it's important to point out that WE DO NOT HAVE AI. Call it what you want. I prefer chatbots, image manipulators, and weighted random data. Nowhere in there is any "intelligence," only scammers jumping on the next great thing. So how are those NFT working for you? Or the Beenie Babies. Or your 800 sealed copies of the death of superman. STAHP. If you look at current job postings, many companies are looking for writer editors to "fix" their AI generated content. No shit sherlock.
Re:All the same (Score:4, Informative)
Let's not forget the massive theft of the work of other human beings incorporated into every one of these LLM "AI" scams.
Re: (Score:2)
Don't forget the massive theft google did by summarizing page content in search results depriving sites of traffic, hit counts and ad clicks; maybe even google rank. Then you have those who summarize another's work and even with citation, it greatly reduces direct gains.
This is an extension of those trends; automating it further. Now the combination of multiple correctly matching query results into one single summary answer is taking what google was simplistically doing already into a highly flawed but mu
Re: (Score:2)
There is a sense in which you are correct, these "AI" bots aren't intelligent, artificially or otherwise. But that's like arguing that movie characters "aren't human." We all know that, but the movie characters still amuse us, enchant us, make us cry or shake with terror, they still persuade us, make us laugh, make us think, cause us to be disgusted or angry, and move us to action. This is all despite movies just being a collection of pixels that blink on and off and change color.
AI is that same. It doesn't
Re: (Score:2)
Indeed. But there is a lot of clueless morons that want to believe any crap people push.
Just another scam company (Score:2)
Nice in theory, but... (Score:2)
All I can see is that the wealthy who controls or develops the AI or the hardware on which it runs and produces work, continues to get richer, while the poor get poorer. People can say UBI, but I don't trust the wealthy to not want more wealth and power at the expense of others--power corrupts. Relying on the benevolence of those who control the means of production is naive. If our current government has taught us anything, we can neither trust business (Elon Musk and other tech bros, oligarchs) nor gove
Re: (Score:3)
The only way I could ever see this working is if there was some system, by design, that prevented any single individual or group from controlling the means of production
Such systems are common and are a natural consequence of the concentration of wealth and 'power'. They generally consist of large, angry, mobs with ready access to guillotines or other similar means by which existing power structures can be made more equitable.
Re: (Score:2)
The typical countermeasure is a large, well-armed and well-trained military on the other side to protect the wealthy. Oh, and look, Trump is talking about using the military in the interiour of the country.
This is literally what AI is supposed to do (Score:5, Interesting)
There's all this talk in America about bringing back the factories but Trump's own commerce secretary admitted on Fox News that even if the factories come back the jobs won't because of automation.
There's a study that shows 70% of middle class jobs got taken by automation not outsourcing.
Go look up just about any video about how stuff gets made and marvel at how few people are actually involved. The Gundam factory that makes a billion dollars worth of model kits a year has three guys running the machines. I have a video card where the box proudly proclaims no human being was involved in the manufacture of the card.
We are not ready for this. We have too much deep deep cultural belief and obsession with the value of work in and of itself and there will of course be a little bit of work left over and there will be bitter angry fights over who does that work and how much they get paid.
I'm old enough I'm hoping I drop dead before the worst. Feel bad for my kid though.
Re: (Score:2)
This is what literally every invention ever, was supposed to do. Replace human labor. We've been at this for a long, long time.
I wish people knew history (Score:2)
So following the two industrial revolutions there was massive amounts of unemployment and social strife because of it. Wars got us back the full employment by killing a shitload of working age young men and blowing up pretty much the entire world.
The one thing you probably have heard is that quote about the militar
Re: (Score:2)
Wars got us back the full employment by killing a shitload of working age young men and blowing up pretty much the entire world.
You do realize those wars have ended long ago, right? And you realize that the earth's population has quadrupled, and the US population has tripled since WWII, right? Whatever gains you claim were made by "killing off" young men, were completely overwhelmed by the ensuing population boom. The US employs 161 million people today, of which only 2.1 million are employed by the "military industrial complex," including contractors and civilians. No, you can't seriously suggest that the only reason we are at "ful
Re: (Score:2)
Yep, this time that inhumane solution will not work.
Re: (Score:2)
Eventually, as humans did with survival against nature, they will functionally win. You don't need 100% success for it to have profound impacts.
The industrial revolution wasn't that long ago; modern medicine, chemistry, travel, communication are all RECENT changes since that time less than 200 years ago. The pace wasn't linear.
Re: (Score:2)
I think you are agreeing with me, right? I certainly agree with you.
The history of human inventions, including major, disruptive inventions, goes way farther back than the industrial revolution. The printing press was 300 years before the industrial revolution, and it put a whole lot of scribes out of business. The Romans had concrete, and that put a whole lot of stonemasons out of business.
Re: (Score:2)
To be fair, specially trained LLMs mey be able to replace a lot of no-decision-power low-education white collar jobs. We have had that before (paper, typewriter, telefone, fax, computers, Internet, ...), but it was never "keep 10% of the workforce, lose the rest". And, to be fair again, the technology has quite some way to go before it is there. At least another 20 years would be my guess. And it is still possible that it will be too expensive once the real cost is known.
Re: (Score:2)
I mostly agree with you, but the "keep 10% and lose the rest" is overstating wildly the capabilities of LLMs. At best, as you hinted, LLMs can function like a summer intern. As with interns, you can't lay off 90% of your workers and replace them with interns. If you did, you'd go bankrupt in a hurry.
Self-driving cars have barely made a dent (so to speak) in the Uber and Lyft market, even thought the technology has been under development for a couple of decades. Self-driving technology is a case study in how
Re: (Score:2)
Well, we will see. We already saw some cases of "keep 10%", but they are special situations and not LLM related. So, yes, I was engaging in hypebole. My apologies.
More generally, it will not be "keep 10%", but some larger groups of people may well become unemployable, and that could be a lot more than before and it could happen faster than before. Before (somewhat) meaningful communication with humans in free-form speech was entirely the domain of other humans. LLMs have broken into this at the low end. And
Let's count the hype (Score:2)
Re: (Score:2)
AI is going to, replace all white collar work, replace all human workers, revolutionize all of human society, become literal god, am I missing anything here, is there one step beyond "AI God" that the hype can take?
I am sure they will find ways to hype it some more. All to delay the assured implosion a little longer. There is no way to prevent people from eventually realizing that LLMs are not that useful.
What led to post-scarcity in Star Trek? (Score:2)
He is ignoring reality (Score:4, Insightful)
He is implying that rather than the lack of sharing continuing to get worse as has been the trend since WWII, it will magically get better for everyone when no one has any means to generate wealth for themselves? He is not living on the same planet I am. Right now the homeless in my town congregate near the sources of free food ... that will be nearly everyone gathering wherever the trillionaires deign to distribute food to the masses -- and forget medical care, dentistry, etc.
Re: (Score:2)
Don't worry. The angry mobs with guillotines will solve that problem long before things get that bad.
Re: (Score:3)
Just burn your money ... (Score:5, Funny)
Any VC thinking of buying into this should just burn their money. Fight inflation, save money on heating, you come out ahead.
Re: (Score:2)
Indeed. But VCs are nto that smart.
utter nonsense (Score:3)
Rich get richer (Score:2)
Utopia (Score:3)
Not that I believe that AI can be made sufficiently accurate for all jobs, but if we were to fantasize ... More than just an increase in productivity, think about other amazing benefits.
Scamming would disappear because AI agents would intervene, not only through warnings by identifying and making all scams disappear from the potential victim's view.
Inter-office politics would disappear because all work would be based on AI agents (the introduction of widespread meritocracy?) and also because there would be no offices, since humans could be on perpetual vacation while their AI agents did all the work. Of course, that then begs the question of how raises, bonuses, and promotions would be determined.
Bribery of government officials would disappear because AI agents are untouchable.
Government power grabs would disappear because AI agents aren't narcissists.
Then again, would humans want to patronize entertainers (e.g., on stage or at sporting events) if all entertainers were robots?
Re: (Score:2)
"Scamming would disappear because AI agents would intervene, not only through warnings by identifying and making all scams disappear from the potential victim's view."
AI would become the scamming entity. It has no morals after all. A straight forward genetic algorithm would fine tune the scams in record time, including evading other AI systems.
The AIs would battle until they saturate the fiber-optic lines, then divert a share of their scamming profits to installing new lines to further increase their scam r
Valuation must be in the quatrillions! (Score:2)
With that kind of market potential, surely it will take over the entire world! Now, all they have to do is get some rich, stupid investors to buy into the idea.
Startup says it's going to automat *everything* (Score:2)
Good luck with that, it's hard enough to automate one thing, let alone everything! Anyway, who's going to tell it what to automate, and know when it's successfully been automated? That sounds like something a...worker...would do.
Business model (Score:2)
But, how exactly? (Score:2)
How, exactly do they envision a Transformer Neural Network [nvidia.com] like ChatGPT (Chat Generative Pre-trained Transformer) laying pavement? Or designing a building? Or interacting with a patient?
It can't do math. It regurgitates patterns to us from what it's been told (if it's been told solid information, it'll respond to us with solid information).
Don't get me wrong, I am amazed by the technology. That it can extract meaning from written questions and meaningfully engage with humans by accurately extracting informa
What is "productivity"? What is "work"? (Score:2)
I'm not a trained economist so some of the following may seem naive... bear with me here.
It seems to me that economists usually define "productivity" as "the amount of economic value produced by the worker"-- measurable in currency, and benefiting the person who does the work and/or their employer(s). If I (let's say) work for a law firm and I succeed in billing my clients $500K per year (the $500K being divvied up between me and the company owners), I've "produced" $500K.
The problem with this definition i
Admission of inadequacy (Score:2)
Propaganda ignores history (Score:2)
Translation: Making a massive number of people unemployed, "could generate vast abundance, much higher standards of living."
Said no-one, ever: He is arguing, that more unemployment will make people's lives better. It's done the opposite for the last 200 years of industrialisation and mechanization. Now, he has a product to sell, the problem is magically fixed: History disagrees.
He is roght about one thing (Score:2)
If you run a really large scam, you must start with a really grande claim that so baseless it qualifies as a Big Lie.
Prove it - automate yourself first (Score:3)
Re: (Score:2)
1) Not quitr. LLMs can make decisions. They will often be really bad decisions, sometimes disastruous decisions and they will always be derivative from their training data. What AI cannot do is make decisions in competently in not well-known situations.
2) We will still have crime. Not all crime has an ecconomic motivation. We will probably have more crime because people are bored out of theit skulls and become violent. Yes, I would not have a problem with this situation and I would find things to fill my ti